[go: up one dir, main page]
More Web Proxy on the site http://driver.im/

CN111402585A - A detection method for occasionally congested paths - Google Patents

A detection method for occasionally congested paths Download PDF

Info

Publication number
CN111402585A
CN111402585A CN202010219735.9A CN202010219735A CN111402585A CN 111402585 A CN111402585 A CN 111402585A CN 202010219735 A CN202010219735 A CN 202010219735A CN 111402585 A CN111402585 A CN 111402585A
Authority
CN
China
Prior art keywords
path
time period
path unit
average speed
candidate
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202010219735.9A
Other languages
Chinese (zh)
Other versions
CN111402585B (en
Inventor
甘志新
陈杰
蔡建南
陈袁芳
邓敏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Central South University
Original Assignee
Central South University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Central South University filed Critical Central South University
Priority to CN202010219735.9A priority Critical patent/CN111402585B/en
Publication of CN111402585A publication Critical patent/CN111402585A/en
Application granted granted Critical
Publication of CN111402585B publication Critical patent/CN111402585B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0125—Traffic data processing
    • G08G1/0129—Traffic data processing for creating historical data or processing based on historical data

Landscapes

  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention provides a method for detecting sporadic congestion paths, which comprises the following steps: acquiring global positioning system GPS track data of a plurality of vehicles in a research area in a target time period; dividing a road network of a research area into a plurality of path units, and determining road direction information of each path unit according to the acquired GPS track data; determining the average speed difference of each path unit in the current time period according to the GPS track data of a plurality of vehicles in the target time period; determining candidate paths according to the topological relation of each path unit in the road network and the average speed difference of each path unit in the current time period; and judging the significance of the candidate route, and determining the candidate route as an accidental congestion route when the significance of the candidate route meets a preset significance level. The method can finish the detection of the accidental congestion path under the condition of considering the road direction, and improve the accuracy of the accidental congestion path detection.

Description

偶发性拥堵路径的探测方法Detection method of occasionally congested path

技术领域technical field

本发明涉及时空数据挖掘与时空统计技术领域，特别涉及一种偶发性拥堵路径的探测方法。The invention relates to the technical field of spatiotemporal data mining and spatiotemporal statistics, in particular to a detection method for an occasional congested path.

背景技术Background technique

随着城市化进程的发展和人民生活水平的不断提高，人均出行需求也随之增加，而相比之下城市道路和基础设施建设的发展速度还远远滞后，直接导致了城市交通的拥堵问题。城市中的交通拥堵问题主要分为两种：一种是经常发生在相似时间发生于相近路径的拥堵现象，称为常发性拥堵(如上下班高峰期)；另一种是指道路交通通行能力由于某些不可控因素所造成的交通拥堵，称为偶发性拥堵(如交通事故)。常发性拥堵的发生位置相对固定，加上及时有效的预防，能够较好的控制常发性拥堵带来的风险。偶发性拥堵是由于大型演出活动或交通事故等突发情况，导致与历史信息相比具有更加显著的拥堵特征，其发生时间与位置通常不可预测，且缺乏历史经验的指导，为此需要交管部门对其做出精准判断，从而避免这种偶发性拥堵所带来的交通影响。因此，准确探测偶发性拥堵对辅助解释偶发性拥堵的成因，进一步加强城市交通管控，最大化地降低二次交通事故都能够提供有效的帮助和指导。With the development of urbanization and the continuous improvement of people's living standards, the per capita travel demand has also increased. In contrast, the development speed of urban roads and infrastructure construction is still far behind, which directly leads to the problem of urban traffic congestion. . Traffic congestion problems in cities are mainly divided into two types: one is the congestion phenomenon that often occurs on similar routes at similar times, called frequent congestion (such as rush hours); the other refers to road traffic capacity. Traffic congestion caused by some uncontrollable factors is called occasional congestion (such as traffic accidents). The location of frequent congestion is relatively fixed, coupled with timely and effective prevention, the risks brought by frequent congestion can be better controlled. Occasional congestion is due to emergencies such as large-scale performances or traffic accidents, resulting in more significant congestion characteristics than historical information. Its occurrence time and location are usually unpredictable and lack the guidance of historical experience. For this reason, traffic control departments are required. Make accurate judgments on it, so as to avoid the traffic impact caused by this occasional congestion. Therefore, accurate detection of occasional congestion can provide effective help and guidance for assisting in explaining the causes of occasional congestion, further strengthening urban traffic control, and maximizing the reduction of secondary traffic accidents.

目前，针对偶发性拥堵探测已发展了一系列方法，根据拥堵特征，可以将现有方法大致分为：(1)基于速度特征的方法，该方法主要是利用路径内车辆通行速度的变化来判定偶发性拥堵。例如，祁宏生等人通过混合高斯分布拟合得到每个路径的速度排名分布函数，进而将具有多峰值且最小峰的概率较小的路径识别为偶发性拥堵路径。(2)基于时间特征的方法，该类方法将偶发性拥堵理解为车辆平均通行时间显著高于历史值的路径。例如，Berk

Figure BDA0002425652290000011
等人给出一种扫描统计方法，依据路径车辆的通行时间来定义似然比函数，首先找出路径通行时间远大于历史平均值的候选区域，将似然比函数作为显著性检验的统计量从候选区域中发现偶发性拥堵的显著时空区域。(3)基于流量-占有率特征的方法，该类方法将下游流量低且占有率的路径定义为偶发性拥堵路径。例如，经典的Mc-Master是基于历史数据绘制流量-占有率二维图，并通过曲线拟合将偶发性拥堵与其他拥堵类型进行区分。(4)基于综合特征的方法，该类方法综合考虑通行速度、道路流量和占有率等多个特征识别偶发性拥堵路径。例如，秦韬等人利用标记有交通状态的历史分类数据训练BP神经网络算法，建立路径的交通状态与通行速度、道路流量以及占有率的映射关系，进而对当前路径的交通状态进行判别。At present, a series of methods have been developed for the detection of occasional congestion. According to the characteristics of congestion, the existing methods can be roughly divided into: (1) The method based on speed characteristics, which mainly uses the change of vehicle speed in the path to determine Occasional congestion. For example, Qi Hongsheng et al. obtained the velocity ranking distribution function of each path by fitting a Gaussian mixture distribution, and then identified a path with multiple peaks and a small probability of the smallest peak as an occasional congested path. (2) Temporal feature-based methods, which understand occasional congestion as a path where the average travel time of vehicles is significantly higher than the historical value. For example, Berk
Figure BDA0002425652290000011
et al. gave a scanning statistical method, which defines the likelihood ratio function according to the transit time of the vehicles on the path. First, find the candidate area where the transit time of the path is much larger than the historical average, and use the likelihood ratio function as the statistic of the significance test. Significant spatiotemporal regions with occasional congestion are found from candidate regions. (3) Methods based on traffic-occupancy characteristics, which define paths with low downstream traffic and occupancy as occasional congested paths. For example, the classic Mc-Master draws a two-dimensional flow-occupancy graph based on historical data, and uses curve fitting to differentiate occasional congestion from other congestion types. (4) Methods based on comprehensive features, which comprehensively consider multiple features such as traffic speed, road flow and occupancy rate to identify occasional congested paths. For example, Qin Tao et al. used the historical classification data marked with the traffic state to train the BP neural network algorithm to establish the mapping relationship between the traffic state of the path and the traffic speed, road flow and occupancy rate, and then discriminate the traffic state of the current path.

通过以上分析可以发现，现有方法虽然能够一定程度地揭示偶发性拥堵发生的路径或时间区间，但均忽略了道路的方向性对挖掘结果的影响，在拥堵的探测过程中忽略道路的方向信息，将会导致局部路径上偶发性拥堵路径的误判和漏判。Through the above analysis, it can be found that although the existing methods can reveal the path or time interval of the occasional congestion to a certain extent, they ignore the influence of the direction of the road on the excavation results, and ignore the direction information of the road in the process of congestion detection. , which will lead to misjudgment and missed judgment of the occasionally congested path on the local path.

发明内容SUMMARY OF THE INVENTION

本发明提供了一种偶发性拥堵路径的探测方法，其目的是为了解决在偶发性拥堵路径的探测过程中忽略道路的方向信息，造成局部路径上偶发性拥堵路径的误判和漏判的问题。The present invention provides a detection method for an occasional congested path, the purpose of which is to solve the problem of ignoring the direction information of the road in the detection process of the occasional congested path, resulting in misjudgment and omission of the occasional congested path on the local path .

为了达到上述目的，本发明的实施例提供了一种偶发性拥堵路径的探测方法，包括：In order to achieve the above object, an embodiment of the present invention provides a method for detecting an occasional congested path, including:

步骤1，获取研究区域内多个车辆在目标时间段内的全球定位系统GPS轨迹数据；所述目标时间段包括当前时间段和历史时间段；Step 1, obtaining GPS track data of multiple vehicles in the research area within the target time period; the target time period includes the current time period and the historical time period;

步骤2，将所述研究区域的路网划分为多个路径单元，并根据获取到的GPS轨迹数据，确定每个路径单元的道路方向信息；Step 2, dividing the road network of the research area into a plurality of path units, and determining the road direction information of each path unit according to the obtained GPS track data;

步骤3，根据所述多个车辆在目标时间段内的GPS轨迹数据，确定每个路径单元在所述当前时间段内的平均速度差异；Step 3, according to the GPS trajectory data of the multiple vehicles in the target time period, determine the average speed difference of each path unit in the current time period;

步骤4，根据所述路网中各路径单元的拓扑关系，以及每个路径单元在所述当前时间段内的平均速度差异，确定出候选路径；Step 4, according to the topological relationship of each path unit in the road network, and the average speed difference of each path unit in the current time period, determine the candidate path;

步骤5，对所述候选路径的显著性进行判别，并当所述候选路径的显著性满足预设的显著性水平时，确定所述候选路径为偶发性拥堵路径。Step 5: Distinguish the significance of the candidate path, and when the significance of the candidate path meets a preset significance level, determine that the candidate path is an occasional congested path.

其中，所述步骤2包括：Wherein, the step 2 includes:

步骤2.1，将所述研究区域的路网等距划分为多个路径单元；Step 2.1, dividing the road network of the research area into multiple path units equidistantly;

步骤2.2，分别针对每个路径单元，根据在所述当前时间段内位于该路径单元上的任一车辆的GPS轨迹数据，确定在所述当前时间段内该车辆的行驶方向，并根据该行驶方向确定该路径单元的道路方向信息。Step 2.2, for each path unit respectively, according to the GPS trajectory data of any vehicle located on the path unit in the current time period, determine the driving direction of the vehicle in the current time period, and according to the driving direction of the vehicle. The direction determines the road direction information for this path element.

其中，所述步骤3包括：Wherein, the step 3 includes:

步骤3.1，获取每个路径单元在所述当前时间段内的当前平均速度；Step 3.1, obtaining the current average speed of each path unit in the current time period;

步骤3.2，获取每个路径单元在所述历史时间段内的历史平均速度；Step 3.2, obtaining the historical average speed of each path unit within the historical time period;

步骤3.3，根据每个路径单元的当前平均速度和历史平均速度，确定每个路径单元在所述当前时间段内的平均速度差异。Step 3.3, according to the current average speed and the historical average speed of each path unit, determine the average speed difference of each path unit in the current time period.

其中，所述步骤3.1包括：Wherein, the step 3.1 includes:

步骤3.11，分别针对多个车辆中的每个车辆，通过线性插值方法获得车辆在所述当前时间段内通过每个路径单元的时间，并根据所述车辆在所述当前时间段内通过每个路径单元的时间，获得所述车辆在所述当前时间段内通过每个路径单元的平均速度；Step 3.11, for each vehicle in the plurality of vehicles, obtain the time for the vehicle to pass through each path unit in the current time period through a linear interpolation method, and according to the current time period for the vehicle to pass through each path unit the time of the path unit, obtaining the average speed of the vehicle passing through each path unit in the current time period;

步骤3.12，分别针对每个路径单元，从所述多个车辆中确定出在所述当前时间段内通过该路径单元的所有车辆，并将所述所有车辆在所述当前时间段内通过该路径单元的平均速度的平均值，作为该路径单元在所述当前时间段内的当前平均速度。Step 3.12: For each route unit, determine all vehicles passing through the route unit in the current time period from the plurality of vehicles, and pass all the vehicles through the route in the current time period The average value of the average speed of the unit, as the current average speed of the path unit in the current time period.

其中，所述步骤3.2包括：Wherein, the step 3.2 includes:

步骤3.21，分别针对多个车辆中的每个车辆，通过线性插值方法获得车辆在所述历史时间段内通过每个路径单元的时间，并根据所述车辆在所述历史时间段内通过每个路径单元的时间，获得所述车辆在所述历史时间段内通过每个路径单元的平均速度；Step 3.21, for each vehicle in the plurality of vehicles, obtain the time that the vehicle passes through each route unit in the historical time period through a linear interpolation method, and according to the the time of the path unit, to obtain the average speed of the vehicle passing through each path unit in the historical time period;

步骤3.22，分别针对每个路径单元，从所述多个车辆中确定出在所述历史时间段内通过该路径单元的所有车辆，并将所述所有车辆在所述历史时间段内通过该路径单元的平均速度的平均值，作为该路径单元在所述历史时间段内的历史平均速度。Step 3.22, for each route unit respectively, determine all vehicles passing through the route unit in the historical time period from the plurality of vehicles, and pass all the vehicles through the route in the historical time period The average value of the average speed of the unit, as the historical average speed of the path unit in the historical time period.

其中，所述步骤3.3包括：Wherein, the step 3.3 includes:

步骤3.31，分别针对每个路径单元，通过公式

Figure BDA0002425652290000041
得到该路径单元在所述当前时间段内的平均速度差异；Step 3.31, for each path unit separately, by formula
Figure BDA0002425652290000041
obtain the average speed difference of the path unit in the current time period;

其中，

Figure BDA0002425652290000042
表示该路径单元在所述当前时间段内的当前平均速度，
Figure BDA0002425652290000043
表示该路径单元在所述历史时间段内的历史平均速度，Δv表示该路径单元在所述当前时间段内的平均速度差异。in,
Figure BDA0002425652290000042
represents the current average speed of the path unit in the current time period,
Figure BDA0002425652290000043
represents the historical average speed of the path unit in the historical time period, and Δv represents the average speed difference of the path unit in the current time period.

其中，所述步骤4包括：Wherein, the step 4 includes:

步骤4.1，根据所述路网中各路径单元的拓扑关系，构建邻接矩阵；Step 4.1, build an adjacency matrix according to the topological relationship of each path unit in the road network;

步骤4.2，将所述路网中各路径单元中平均速度差异小于0的路径单元作为候选种子单元；Step 4.2, taking the path unit whose average speed difference is less than 0 in each path unit in the road network as a candidate seed unit;

步骤4.3，分别针对每个候选种子单元，执行如下步骤：Step 4.3, for each candidate seed unit, perform the following steps:

按照所述路网中各路径单元的拓扑关系向一阶邻域扩展，计算候选种子单元和每一个邻近路径单元的局部Gi *指数，并选择计算出的局部Gi *指数中绝对值最大的一个邻近路径单元与该候选种子单元进行合并，直至所有的一阶邻近路径单元合并完成或者计算出的局部Gi *指数的绝对值不再增大为止，得到合并路径；Expand to the first-order neighborhood according to the topological relationship of each path unit in the road network, calculate the local G i * index of the candidate seed unit and each adjacent path unit, and select the calculated local G i * index with the largest absolute value One adjacent path unit of , and the candidate seed unit are merged, until all the first-order adjacent path units are merged or the absolute value of the calculated local G i * index no longer increases, and the merged path is obtained;

按照所述路网中各路径单元的拓扑关系，扩展合并路径的k阶邻近路径单元，直至计算出的k阶邻近路径单元的局部Gi *指数的绝对值不再增大为止，得到候选路径；其中k为大于或等于2的整数。According to the topological relationship of each path unit in the road network, expand the k-order adjacent path units of the merged path until the calculated absolute value of the local G i * index of the k-order adjacent path units no longer increases, and obtain a candidate path ; where k is an integer greater than or equal to 2.

其中，所述局部Gi *指数的计算公式为：Wherein, the calculation formula of the local G i * index is:

Figure BDA0002425652290000044
Figure BDA0002425652290000044

其中，Δvj表示路径单元j的历史平均速度与当前平均速度之间的差值，

Figure BDA0002425652290000045
表示所述研究区域内所有平均速度差异的平均值，n表示所述研究区域内路径单元的总数，wi,j为路径单元i和路径单元j的邻接矩阵，s为所述研究区域的方差。where Δv j represents the difference between the historical average speed of path unit j and the current average speed,
Figure BDA0002425652290000045
represents the average value of all average speed differences in the study area, n represents the total number of path units in the study area, w i,j is the adjacency matrix of path unit i and path unit j, s is the variance of the study area .

其中，s的计算公式为：Among them, the calculation formula of s is:

Figure BDA0002425652290000051
Figure BDA0002425652290000051

其中，Δvj表示路径单元j的历史平均速度与当前平均速度之间的差值，

Figure BDA0002425652290000052
表示所述研究区域内所有平均速度差异的平均值，n表示所述研究区域内路径单元的总数。where Δv j represents the difference between the historical average speed of path unit j and the current average speed,
Figure BDA0002425652290000052
represents the average of all mean speed differences in the study area, and n represents the total number of path units in the study area.

其中，所述步骤5包括：Wherein, the step 5 includes:

步骤5.1，分别针对每个候选路径，执行如下步骤：Step 5.1, for each candidate path, perform the following steps:

对候选路径生成N个模拟数据集；Generate N simulated datasets for candidate paths;

计算每个模拟数据集中所述候选路径的似然比统计量得分LLRobs；Calculate the likelihood ratio statistic score LLR obs of the candidate paths in each simulated dataset;

通过公式

Figure BDA0002425652290000053
计算得到所述候选路径的显著性；其中，#(fi)表示符合条件fi的个数，N为模拟数据集的总个数，LLRres为真实数据中所述候选路径的似然比统计量得分，Si表示所述候选路径，p_value(Si)表示所述候选路径的显著性；by formula
Figure BDA0002425652290000053
Calculate the saliency of the candidate path; wherein #(f i ) represents the number of eligible f i , N is the total number of simulated data sets, and LLR res is the likelihood ratio of the candidate path in the real data statistic score, S i represents the candidate path, p_value(S i ) represents the significance of the candidate path;

当所述候选路径的显著性p_value(Si)≤α时，确定所述候选路径的显著性满足预设的显著性水平，并确定所述候选路径为偶发性拥堵路径；其中，α为预设的显著性水平。When the significance p_value(S i ) of the candidate path is less than or equal to α, determine that the significance of the candidate path satisfies a preset significance level, and determine that the candidate path is an occasional congestion path; set significance level.

本发明的上述方案至少有如下的有益效果：The above-mentioned scheme of the present invention has at least the following beneficial effects:

在本发明的实施例中，通过将研究区域的路网划分为多个路径单元，根据研究区域内车辆的GPS轨迹数据，确定每个路径单元的道路方向信息以及每个路径单元在当前时间段内的平均速度差异，然后根据各路径单元的拓扑关系以及每个路径单元在当前时间段内的平均速度差异，从多个路径单元中确定出候选路径，最终对候选路径的显著性进行判别，并当候选路径的显著性满足预设的显著性水平时，确定候选路径为偶发性拥堵路径。其中由于每个路径单元包含道路方向信息，因而最终确定的偶发性拥堵路径也包含道路方向信息，进而实现了在顾及道路方向的情况下完成偶发性拥堵路径的探测的效果，提升了偶发性拥堵路径探测的准确性，提高了交通管理部门辅助解决城市交通拥堵问题的实用性与可靠性。In the embodiment of the present invention, by dividing the road network of the study area into multiple path units, the road direction information of each path unit and the current time period of each path unit are determined according to the GPS trajectory data of the vehicles in the study area. Then according to the topological relationship of each path unit and the average speed difference of each path unit in the current time period, the candidate path is determined from multiple path units, and finally the significance of the candidate path is judged, And when the significance of the candidate path satisfies the preset significance level, the candidate path is determined to be an occasional congested path. Since each path unit contains road direction information, the finally determined occasionally congested path also contains road direction information, thereby realizing the effect of completing the detection of the occasionally congested path under the condition of considering the road direction, and improving the occasional congestion The accuracy of the path detection improves the practicability and reliability of the traffic management department to assist in solving the problem of urban traffic congestion.

附图说明Description of drawings

为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图示出的结构获得其他的附图。In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present invention, and for those of ordinary skill in the art, other drawings can also be obtained according to the structures shown in these drawings without creative efforts.

图1是本发明实施例的偶发性拥堵路径的探测方法的流程图。FIG. 1 is a flowchart of a method for detecting an occasional congested path according to an embodiment of the present invention.

具体实施方式Detailed ways

下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

如图1所示，本发明的实施例提供了一种偶发性拥堵路径的探测方法，该探测方法包括如下步骤：As shown in FIG. 1 , an embodiment of the present invention provides a detection method for an occasional congested path, and the detection method includes the following steps:

步骤1，获取研究区域内多个车辆在目标时间段内的全球定位系统GPS轨迹数据；所述目标时间段包括当前时间段和历史时间段。Step 1: Obtain GPS track data of multiple vehicles in the research area within a target time period; the target time period includes a current time period and a historical time period.

其中，在本发明的实施例中，上述研究区域可以为任一地理区域，如深圳市某市区；上述目标时间段可以为任一指定时间段(该时间段包含发生过偶发性拥堵的时间段)，如2012年1月份每个工作日的19:30-20:30；上述当前时间段为目标时间段内发生偶发性拥堵的时间段，历史时间段是指目标时间段内除当前时间段以外的其他时间段。举例说明，假设目标时间段为2012年1月份每个工作日的19:30-20:30，将1月15日的19:30-20:30作为当前时间段(鉴于1月15日20:00举办的演唱会活动造成了偶发性拥堵)，其余工作日的19:30-20:30为历史时间段。Wherein, in the embodiment of the present invention, the above-mentioned research area can be any geographical area, such as an urban area in Shenzhen; the above-mentioned target time period can be any specified time period (the time period includes the time when occasional congestion occurs period), such as 19:30-20:30 every working day in January 2012; the above current time period is the time period when occasional congestion occurs within the target time period, and the historical time period refers to the target time period except the current time period other than the period. For example, suppose the target time period is 19:30-20:30 every working day in January 2012, and the current time period is 19:30-20:30 on January 15th (given that on January 15th 20:30: 00 caused occasional congestion), and 19:30-20:30 on other working days is a historical time period.

具体的，在本发明的实施例中，可通过从相关设备提取GPS(全球定位系统)轨迹数据的方式，获得研究区域内多个车辆在目标时间段内的GPS轨迹数据，即，每个车辆在目标时间段内的GPS轨迹数据。其中，上述车辆可以为出租车、公交车等任一车型的车辆。Specifically, in the embodiment of the present invention, the GPS trajectory data of multiple vehicles in the research area in the target time period can be obtained by extracting GPS (Global Positioning System) trajectory data from related equipment, that is, each vehicle GPS trajectory data during the target time period. The above-mentioned vehicle may be a vehicle of any type, such as a taxi or a bus.

其中，上述车辆在目标时间段内的GPS轨迹数据包括该车辆的多个轨迹点，为提高探测偶发性拥堵路径的准确性，在获取到车辆在目标时间段内的GPS轨迹数据后，会剔除GPS轨迹数据中的异常点(如行程短于预设值的轨迹点)。Among them, the GPS trajectory data of the vehicle in the target time period includes multiple trajectory points of the vehicle. In order to improve the accuracy of detecting the occasionally congested path, after the GPS trajectory data of the vehicle in the target time period is obtained, it will be eliminated. Abnormal points in GPS trajectory data (such as trajectory points whose travel is shorter than the preset value).

步骤2，将所述研究区域的路网划分为多个路径单元，并根据获取到的GPS轨迹数据，确定每个路径单元的道路方向信息。Step 2: Divide the road network of the research area into a plurality of path units, and determine the road direction information of each path unit according to the acquired GPS track data.

其中，在本发明的实施例中，上述步骤2的具体实现方式包括如下步骤：Wherein, in the embodiment of the present invention, the specific implementation of the above step 2 includes the following steps:

步骤2.1，将所述研究区域的路网等距划分为多个路径单元。Step 2.1: Divide the road network of the study area into multiple path units equidistantly.

其中，在本发明的实施例中，在得到多个路径单元后，会将之前得到的车辆的GPS轨迹数据中的轨迹点和路网进行匹配，以保证所统计的每一个轨迹点都能在路网上。Among them, in the embodiment of the present invention, after obtaining multiple path units, the track points in the GPS track data of the vehicle obtained before will be matched with the road network, so as to ensure that each track point can be road network.

步骤2.2，分别针对每个路径单元，根据在所述当前时间段内位于该路径单元上的任一车辆的GPS轨迹数据，确定在所述当前时间段内该车辆的行驶方向，并根据该行驶方向确定该路径单元的道路方向信息。Step 2.2, for each path unit respectively, according to the GPS trajectory data of any vehicle located on the path unit in the current time period, determine the driving direction of the vehicle in the current time period, and according to the driving direction of the vehicle. The direction determines the road direction information for this path element.

具体的，在本发明的实施例中，针对每个路径单元，会根据在当前时间段内位于该路径单元上的同一车辆的GPS轨迹数据(如前后采点位置的变化)，确定在当前时间段内该车辆的行驶方向，并将该行驶方向作为该路径单元的道路方向信息。Specifically, in the embodiment of the present invention, for each path unit, the current time will be determined according to the GPS trajectory data of the same vehicle located on the path unit in the current time period (such as the change of the location of the points before and after) The driving direction of the vehicle in the segment is taken as the road direction information of the path unit.

步骤3，根据所述多个车辆在目标时间段内的GPS轨迹数据，确定每个路径单元在所述当前时间段内的平均速度差异。Step 3, according to the GPS trajectory data of the multiple vehicles in the target time period, determine the average speed difference of each path unit in the current time period.

其中，在本发明的实施例中，上述步骤3的具体实现方式包括如下步骤：Wherein, in the embodiment of the present invention, the specific implementation of the above step 3 includes the following steps:

步骤3.1，获取每个路径单元在所述当前时间段内的当前平均速度。Step 3.1, obtaining the current average speed of each path unit in the current time period.

步骤3.2，获取每个路径单元在所述历史时间段内的历史平均速度。Step 3.2: Obtain the historical average speed of each path unit in the historical time period.

步骤3.3，根据每个路径单元的当前平均速度和历史平均速度，确定每个路径单元在所述当前时间段内的平均速度差异。Step 3.3, according to the current average speed and the historical average speed of each path unit, determine the average speed difference of each path unit in the current time period.

具体的，上述步骤3.1的具体实现方式包括如下步骤：Specifically, the specific implementation of the above step 3.1 includes the following steps:

步骤3.11，分别针对多个车辆中的每个车辆，通过线性插值方法获得车辆在所述当前时间段内通过每个路径单元的时间，并根据所述车辆在所述当前时间段内通过每个路径单元的时间，获得所述车辆在所述当前时间段内通过每个路径单元的平均速度。Step 3.11, for each vehicle in the plurality of vehicles, obtain the time for the vehicle to pass through each path unit in the current time period through a linear interpolation method, and according to the current time period for the vehicle to pass through each path unit The time of the path unit, the average speed of the vehicle passing through each path unit in the current time period is obtained.

具体的，针对每个车辆，可通过插值公式

Figure BDA0002425652290000081
获得该车辆在当前时间段内通过每个路径单元的时间。其中，ti(i＝1,...,n)表示该车辆在当前时间段内通过每个原始轨迹点(即GPS轨迹数据中的轨迹点)的时刻，Ti(i＝1,…,n)表示该车辆在当前时间段内通过每个路径单元的时间，li1表示该车辆在当前时间段内与Ti时刻最近的前一个轨迹点的距离，li2表示该车辆在当前时间段内与Ti时刻最近的后一个轨迹点的距离。Specifically, for each vehicle, the interpolation formula can be used
Figure BDA0002425652290000081
Get the time that the vehicle passes through each path unit in the current time period. Among them, t i (i=1,...,n) represents the moment when the vehicle passes through each original trajectory point (that is, the trajectory point in the GPS trajectory data) in the current time period, and T i (i=1,... ,n) represents the time that the vehicle passes through each path unit in the current time period, l i1 represents the distance between the vehicle and the previous trajectory point closest to time T i in the current time period, and l i2 represents the vehicle at the current time The distance to the next trajectory point closest to time Ti in the segment.

需要说明的是，由于在对路网进行划分时便清楚每个路径单元的长度，因此在得到车辆在当前时间段内通过每个路径单元的时间后，便能获得该车辆在当前时间段内通过每个路径单元的平均速度。It should be noted that, since the length of each path unit is clear when the road network is divided, after obtaining the time that the vehicle passes through each path unit in the current time period, the vehicle in the current time period can be obtained. Average speed through each path unit.

步骤3.12，分别针对每个路径单元，从所述多个车辆中确定出在所述当前时间段内通过该路径单元的所有车辆，并将所述所有车辆在所述当前时间段内通过该路径单元的平均速度的平均值，作为该路径单元在所述当前时间段内的当前平均速度。Step 3.12: For each route unit, determine all vehicles passing through the route unit in the current time period from the plurality of vehicles, and pass all the vehicles through the route in the current time period The average value of the average speed of the unit, as the current average speed of the path unit in the current time period.

其中，上述步骤3.2的具体实现方式包括如下步骤：The specific implementation of the above step 3.2 includes the following steps:

步骤3.21，分别针对多个车辆中的每个车辆，通过线性插值方法获得车辆在所述历史时间段内通过每个路径单元的时间，并根据所述车辆在所述历史时间段内通过每个路径单元的时间，获得所述车辆在所述历史时间段内通过每个路径单元的平均速度。Step 3.21, for each vehicle in the plurality of vehicles, obtain the time that the vehicle passes through each route unit in the historical time period through a linear interpolation method, and according to the The time of the path unit, the average speed of the vehicle passing through each path unit in the historical time period is obtained.

具体的，针对每个车辆，可通过插值公式

Figure BDA0002425652290000082
获得该车辆在历史时间段内通过每个路径单元的时间。其中，ti(i＝1,...,n)表示该车辆在历史时间段内通过每个原始轨迹点(即GPS轨迹数据中的轨迹点)的时刻，Ti(i＝1,…,n)表示该车辆在历史时间段内通过每个路径单元的时间，li1表示该车辆在历史时间段内与Ti时刻最近的前一个轨迹点的距离，li2表示该车辆在历史时间段内与Ti时刻最近的后一个轨迹点的距离。Specifically, for each vehicle, the interpolation formula can be used
Figure BDA0002425652290000082
Get the time that the vehicle passed through each route unit in the historical time period. Among them, t i (i=1,...,n) represents the moment when the vehicle passes through each original trajectory point (that is, the trajectory point in the GPS trajectory data) in the historical time period, and T i (i=1,... ,n) represents the time that the vehicle passes through each path unit in the historical time period, l i1 represents the distance between the vehicle and the previous trajectory point closest to time T i in the historical time period, and l i2 represents the vehicle in the historical time period The distance to the next trajectory point closest to time Ti in the segment.

需要说明的是，由于在对路网进行划分时便清楚每个路径单元的长度，因此在得到车辆在历史时间段内通过每个路径单元的时间后，便能获得该车辆在历史时间段内通过每个路径单元的平均速度。It should be noted that since the length of each path unit is clear when the road network is divided, after obtaining the time that the vehicle passes through each path unit in the historical time period, the vehicle in the historical time period can be obtained. Average speed through each path unit.

步骤3.22，分别针对每个路径单元，从所述多个车辆中确定出在所述历史时间段内通过该路径单元的所有车辆，并将所述所有车辆在所述历史时间段内通过该路径单元的平均速度的平均值，作为该路径单元在所述历史时间段内的历史平均速度。Step 3.22, for each route unit respectively, determine all vehicles passing through the route unit in the historical time period from the plurality of vehicles, and pass all the vehicles through the route in the historical time period The average value of the average speed of the unit, as the historical average speed of the path unit in the historical time period.

其中，上述步骤3.3的具体实现方式包括如下步骤：The specific implementation of the above step 3.3 includes the following steps:

步骤3.31，分别针对每个路径单元，通过公式

Figure BDA0002425652290000091
得到该路径单元在所述当前时间段内的平均速度差异。Step 3.31, for each path unit separately, by formula
Figure BDA0002425652290000091
Obtain the average speed difference of the path unit in the current time period.

其中，

Figure BDA0002425652290000092
表示该路径单元在所述当前时间段内的当前平均速度，
Figure BDA0002425652290000093
表示该路径单元在所述历史时间段内的历史平均速度，Δv表示该路径单元在所述当前时间段内的平均速度差异。in,
Figure BDA0002425652290000092
represents the current average speed of the path unit in the current time period,
Figure BDA0002425652290000093
represents the historical average speed of the path unit in the historical time period, and Δv represents the average speed difference of the path unit in the current time period.

步骤4，根据所述路网中各路径单元的拓扑关系，以及每个路径单元在所述当前时间段内的平均速度差异，确定出候选路径。Step 4: Determine a candidate path according to the topological relationship of each path unit in the road network and the average speed difference of each path unit in the current time period.

其中，在本发明的实施例中，通过以空间局部自相关指数作为偶发性拥堵路径探测过程的优化函数，先在候选种子单元的一阶邻域内逐个测试，使得每次扩展结果的空间局部自相关指数值不断增大，进而对合并后路径的下一阶邻域进行扩展。Among them, in the embodiment of the present invention, the spatial local autocorrelation index is used as the optimization function of the detection process of the occasional congested path, and the first-order neighborhood of the candidate seed unit is tested one by one, so that the spatial local autocorrelation of each expansion result is obtained. The value of the correlation index increases continuously, and then the next-order neighborhood of the merged path is expanded.

具体的，上述步骤4的具体实现方式包括如下步骤：Specifically, the specific implementation of the above step 4 includes the following steps:

步骤4.1，根据所述路网中各路径单元的拓扑关系，构建邻接矩阵。Step 4.1, construct an adjacency matrix according to the topological relationship of each path unit in the road network.

其中，在构建邻接矩阵时，将具有拓扑连接的相邻路径单元的邻接值设置为1，其余均设置为0。Among them, when constructing the adjacency matrix, the adjacency value of the adjacent path units with topological connections is set to 1, and the rest are set to 0.

步骤4.2，将所述路网中各路径单元中平均速度差异小于0的路径单元作为候选种子单元。Step 4.2, taking the path unit whose average speed difference is less than 0 among the path units in the road network as the candidate seed unit.

步骤4.3，分别针对每个候选种子单元，执行如下步骤：Step 4.3, for each candidate seed unit, perform the following steps:

首先按照所述路网中各路径单元的拓扑关系向一阶邻域扩展，计算候选种子单元和每一个邻近路径单元的局部Gi *指数，并选择计算出的局部Gi *指数中绝对值最大的一个邻近路径单元与该候选种子单元进行合并，直至所有的一阶邻近路径单元合并完成或者计算出的局部Gi *指数的绝对值不再增大为止，得到合并路径；然后按照所述路网中各路径单元的拓扑关系，扩展合并路径的k阶邻近路径单元，直至计算出的k阶邻近路径单元的局部Gi *指数的绝对值不再增大为止，得到候选路径；其中k为大于或等于2的整数。Firstly, according to the topological relationship of each path unit in the road network, expand to the first-order neighborhood, calculate the local G i * index of the candidate seed unit and each adjacent path unit, and select the absolute value of the calculated local G i * index The largest adjacent path unit is merged with the candidate seed unit, until all the first-order adjacent path units are merged or the absolute value of the calculated local G i * index no longer increases, and the merged path is obtained; then follow the described The topological relationship of each path unit in the road network, expand the k-order adjacent path unit of the merged path, until the absolute value of the calculated local G i * index of the k-order adjacent path unit no longer increases, get the candidate path; where k is an integer greater than or equal to 2.

其中，上述局部Gi *指数的计算公式为：Among them, the calculation formula of the above local G i * index is:

Figure BDA0002425652290000101
Figure BDA0002425652290000101

其中，Δvj表示路径单元j的历史平均速度与当前平均速度之间的差值，

Figure BDA0002425652290000102
表示所述研究区域内所有平均速度差异的平均值，n表示所述研究区域内路径单元的总数，wi,j为路径单元i和路径单元j的邻接矩阵，s为所述研究区域的方差。需要说明的是，上述局部Gi *指数的计算公式可用于计算上述步骤4.3中的各局部Gi *指数，举例说明，当计算候选种子单元(记为A)和某一邻近路径单元(记为B)的局部Gi *指数时，wi,j为A和路径单元B的邻接矩阵。where Δv j represents the difference between the historical average speed of path unit j and the current average speed,
Figure BDA0002425652290000102
represents the average value of all average speed differences in the study area, n represents the total number of path units in the study area, w i,j is the adjacency matrix of path unit i and path unit j, s is the variance of the study area . It should be noted that the above calculation formula of the local G i * index can be used to calculate each local G i * index in the above step 4.3. For example, when calculating the candidate seed unit (denoted as A) and a certain adjacent path unit (denoted as A) is the local G i * index of B), w i,j is the adjacency matrix of A and path element B.

其中，s的计算公式为：Among them, the calculation formula of s is:

Figure BDA0002425652290000103
Figure BDA0002425652290000103

其中，Δvj表示路径单元j的历史平均速度与当前平均速度之间的差值，

Figure BDA0002425652290000104
表示所述研究区域内所有平均速度差异的平均值，n表示所述研究区域内路径单元的总数。where Δv j represents the difference between the historical average speed of path unit j and the current average speed,
Figure BDA0002425652290000104
represents the average of all mean speed differences in the study area, and n represents the total number of path units in the study area.

步骤5，对所述候选路径的显著性进行判别，并当所述候选路径的显著性满足预设的显著性水平时，确定所述候选路径为偶发性拥堵路径。Step 5: Distinguish the significance of the candidate path, and when the significance of the candidate path meets a preset significance level, determine that the candidate path is an occasional congested path.

其中，在本发明的实施例中，上述步骤5的具体实现方式包括如下步骤：Wherein, in the embodiment of the present invention, the specific implementation of the above step 5 includes the following steps:

步骤5.1，分别针对每个候选路径，执行如下步骤：Step 5.1, for each candidate path, perform the following steps:

第一步，对候选路径生成N个模拟数据集。In the first step, N simulated datasets are generated for the candidate paths.

其中，每个模拟数据集为该候选路径的平均速度，其服从λ为历史平均速度的泊松分布。Among them, each simulated data set is the average speed of the candidate path, which obeys a Poisson distribution with λ being the historical average speed.

第二步，计算每个模拟数据集中所述候选路径的似然比统计量得分LLRobs。In the second step, the likelihood ratio statistic score LLR obs of the candidate paths in each simulated dataset is calculated.

第三步，通过公式

Figure BDA0002425652290000111
计算得到所述候选路径的显著性；其中，#(fi)表示符合条件fi的个数，N为模拟数据集的总个数，LLRres为真实数据中所述候选路径的似然比统计量得分，Si表示所述候选路径，p_value(Si)表示所述候选路径的显著性。其中，当所述候选路径的显著性p_value(Si)≤α时，确定所述候选路径的显著性满足预设的显著性水平，并确定所述候选路径为偶发性拥堵路径；其中，α为预设的显著性水平。The third step, through the formula
Figure BDA0002425652290000111
Calculate the saliency of the candidate path; wherein #(f i ) represents the number of eligible f i , N is the total number of simulated data sets, and LLR res is the likelihood ratio of the candidate path in the real data Statistical score, S i represents the candidate path, p_value(S i ) represents the significance of the candidate path. Wherein, when the significance p_value(S i ) of the candidate path is less than or equal to α, it is determined that the significance of the candidate path satisfies a preset significance level, and the candidate path is determined to be an occasional congested path; where α is the preset significance level.

需要说明的是，候选路径的似然比统计量得分的计算公式为：It should be noted that the calculation formula of the likelihood ratio statistic score of the candidate path is:

Figure BDA0002425652290000112
Figure BDA0002425652290000112

其中，Si表示候选路径，v为候选路径的平均速度，

Figure BDA0002425652290000113
为候选路径的历史平均速度。需要进一步说明的是，在计算每个模拟数据集中所述候选路径的似然比统计量得分LLRobs时，用的是模拟数据集中的数据，而在计算LLRres时，用的是真实数据(如上述步骤3中的数据)。where S i represents the candidate path, v is the average speed of the candidate path,
Figure BDA0002425652290000113
is the historical average speed of the candidate path. It should be further noted that when calculating the likelihood ratio statistic score LLR obs of the candidate paths in each simulated data set, the data in the simulated data set is used, and when calculating the LLR res , the real data ( data in step 3 above).

值得一提的是，在本发明的实施例中，通过将研究区域的路网划分为多个路径单元，根据研究区域内车辆的GPS轨迹数据，确定每个路径单元的道路方向信息以及每个路径单元在当前时间段内的平均速度差异，然后根据各路径单元的拓扑关系以及每个路径单元在当前时间段内的平均速度差异，从多个路径单元中确定出候选路径，最终对候选路径的显著性进行判别，并当候选路径的显著性满足预设的显著性水平时，确定候选路径为偶发性拥堵路径。其中由于每个路径单元包含道路方向信息，因而最终确定的偶发性拥堵路径也包含道路方向信息，进而实现了在顾及道路方向的情况下完成偶发性拥堵路径的探测的效果，提升了偶发性拥堵路径探测的准确性，提高了交通管理部门辅助解决城市交通拥堵问题的实用性与可靠性。It is worth mentioning that, in the embodiment of the present invention, by dividing the road network of the study area into multiple path units, the road direction information of each path unit and each path unit are determined according to the GPS trajectory data of vehicles in the study area. The average speed difference of the path units in the current time period, and then according to the topological relationship of each path unit and the average speed difference of each path unit in the current time period, the candidate path is determined from multiple path units, and finally the candidate path is determined. The significance of the candidate path is determined, and when the significance of the candidate path meets the preset significance level, the candidate path is determined to be an occasional congested path. Since each path unit contains road direction information, the finally determined occasionally congested path also contains road direction information, thereby realizing the effect of completing the detection of the occasionally congested path under the condition of considering the road direction, and improving the occasional congestion The accuracy of the path detection improves the practicability and reliability of the traffic management department to assist in solving the problem of urban traffic congestion.

接下来，采用深圳市某市区2012年1月份工作日的出租车GPS轨迹数据对本发明的具体实施进行说明，下面将结合此实例具体说明本发明探测城市偶发性拥堵路径的具体实施步骤：Next, the specific implementation of the present invention will be described by using the taxi GPS trajectory data of a certain urban area in Shenzhen in January 2012. The specific implementation steps of the present invention to detect the urban occasional congestion path will be described below in conjunction with this example:

1)首先提取2012年1月工作日19:30-20:30内的出租车GPS轨迹数据，将1月15日的数据作为当前时间段的检测数据(鉴于当天20:00举办的演唱会活动造成的偶发性拥堵)，其余工作日的数据作为历史数据，清理GPS轨迹数据中的异常点，将提取得到的轨迹点和城市路网匹配起来。1) First extract the GPS trajectory data of taxis from 19:30-20:30 on working days in January 2012, and use the data on January 15 as the detection data of the current time period (in view of the concert held at 20:00 on the same day) The data of the remaining working days are used as historical data, the abnormal points in the GPS trajectory data are cleaned up, and the extracted trajectory points are matched with the urban road network.

2)将该市区的路网划分成100米的路径单元，根据路径单元上轨迹的行驶方向确定道路的方向性。2) The urban road network is divided into 100-meter path units, and the directionality of the road is determined according to the traveling direction of the trajectory on the path unit.

3)以10分钟为一个时间区间，利用线性插值方法计算每节路径单元节点上车辆通过的时间，得到路径单元的平均速度，计算每节路径单元中当前时间段内平均速度的差异。3) Taking 10 minutes as a time interval, use the linear interpolation method to calculate the passing time of vehicles on each path element node, obtain the average speed of the path element, and calculate the difference of the average speed in the current time period in each path element.

4)根据每节路径单元的拓扑关系构建邻接矩阵，计算每节路径单元的局部Gi *指数，根据多向扩展策略得到候选路径(即候选的偶发性拥堵路径)。4) Construct an adjacency matrix according to the topological relationship of each path unit, calculate the local G i * index of each path unit, and obtain a candidate path (ie, a candidate occasionally congested path) according to the multidirectional expansion strategy.

5)在模拟数据集中，每套模拟数据集中路径单元的平均速度服从λ为历史平均速度的泊松分布，计算候选偶发性拥堵路径的似然比得分。5) In the simulation data set, the average speed of the path units in each simulation data set obeys the Poisson distribution with λ as the historical average speed, and calculates the likelihood ratio score of the candidate occasionally congested path.

6)计算候选偶发性拥堵路径的p值(即显著性值)时(其中，预设的显著性水平设置为0.05)，若候选的偶发性拥堵路径的p值低于0.05，则被识别为偶发性拥堵路径。6) When calculating the p-value (ie, the significance value) of the candidate occasionally congested path (wherein, the preset significance level is set to 0.05), if the p-value of the candidate occasionally congested path is lower than 0.05, it is identified as Occasional congested paths.

以上所述是本发明的优选实施方式，应当指出，对于本技术领域的普通技术人员来说，在不脱离本发明所述原理的前提下，还可以作出若干改进和润饰，这些改进和润饰也应视为本发明的保护范围。The above are the preferred embodiments of the present invention. It should be pointed out that for those skilled in the art, without departing from the principles of the present invention, several improvements and modifications can be made. It should be regarded as the protection scope of the present invention.

Claims (10)

1. A method for detecting sporadic congestion paths, comprising:
step 1, acquiring global positioning system GPS track data of a plurality of vehicles in a research area in a target time period; the target time period comprises a current time period and a historical time period;
step 2, dividing the road network of the research area into a plurality of path units, and determining road direction information of each path unit according to the acquired GPS track data;
step 3, determining the average speed difference of each path unit in the current time period according to the GPS track data of the vehicles in the target time period;
step 4, determining candidate paths according to the topological relation of each path unit in the road network and the average speed difference of each path unit in the current time period;
and 5, judging the significance of the candidate route, and determining the candidate route as an occasional congestion route when the significance of the candidate route meets a preset significance level.
2. The detection method according to claim 1, wherein the step 2 comprises:
step 2.1, equally dividing the road network of the research area into a plurality of path units;
and 2.2, respectively aiming at each path unit, determining the driving direction of the vehicle in the current time period according to the GPS track data of any vehicle positioned on the path unit in the current time period, and determining the road direction information of the path unit according to the driving direction.
3. The detection method according to claim 1, wherein the step 3 comprises:
step 3.1, acquiring the current average speed of each path unit in the current time period;
step 3.2, acquiring the historical average speed of each path unit in the historical time period;
and 3.3, determining the average speed difference of each path unit in the current time period according to the current average speed and the historical average speed of each path unit.
4. A detection method according to claim 3, characterised in that said step 3.1 comprises:
step 3.11, respectively aiming at each vehicle in the plurality of vehicles, obtaining the time of the vehicle passing through each path unit in the current time period by a linear interpolation method, and obtaining the average speed of the vehicle passing through each path unit in the current time period according to the time of the vehicle passing through each path unit in the current time period;
and 3.12, respectively aiming at each path unit, determining all vehicles which pass through the path unit in the current time period from the plurality of vehicles, and taking the average value of the average speeds of all vehicles which pass through the path unit in the current time period as the current average speed of the path unit in the current time period.
5. A detection method according to claim 3, characterised in that said step 3.2 comprises:
step 3.21, respectively aiming at each vehicle in the plurality of vehicles, obtaining the time of the vehicle passing through each path unit in the historical time period by a linear interpolation method, and obtaining the average speed of the vehicle passing through each path unit in the historical time period according to the time of the vehicle passing through each path unit in the historical time period;
and 3.22, respectively aiming at each path unit, determining all vehicles which pass through the path unit in the historical time period from the plurality of vehicles, and taking the average value of the average speeds of all vehicles which pass through the path unit in the historical time period as the historical average speed of the path unit in the historical time period.
6. A detection method according to claim 3, characterised in that said step 3.3 comprises:
step 3.31, respectively aiming at each path unit, passing through a formula
Figure FDA0002425652280000021
Obtaining the average speed difference of the path unit in the current time period;
wherein,
Figure FDA0002425652280000022
representing the current average speed of the path element over said current time period,
Figure FDA0002425652280000023
representing the history of the path unit in the history time periodThe average speed, Δ v, represents the average speed difference of the path unit over the current time period.
7. The detection method according to claim 1, wherein the step 4 comprises:
step 4.1, constructing an adjacency matrix according to the topological relation of each path unit in the road network;
step 4.2, taking the path units with the average speed difference smaller than 0 in each path unit in the road network as candidate seed units;
step 4.3, respectively aiming at each candidate seed unit, executing the following steps:
expanding to the first-order neighborhood according to the topological relation of each path unit in the road network, and calculating the local G of the candidate seed unit and each adjacent path uniti *Index and select the calculated local Gi *Combining the adjacent path unit with the maximum absolute value in the exponent with the candidate seed unit until all the first-order adjacent path units are combined or the calculated local G is obtainedi *Obtaining a merging path until the absolute value of the exponent is not increased;
expanding k-order adjacent path units of the merged path according to the topological relation of each path unit in the road network until the local G of the k-order adjacent path units is calculatedi *Obtaining a candidate path until the absolute value of the index is not increased; wherein k is an integer greater than or equal to 2.
8. The detection method according to claim 7, wherein the local Gi index is calculated by the formula:
Figure FDA0002425652280000031
wherein, Δ vjRepresenting the difference between the historical average speed and the current average speed of path element j,
Figure FDA0002425652280000032
representing the average of all average velocity differences within said investigation region, n representing the total number of path elements within said investigation region, wi,jIs the adjacency matrix of path element i and path element j, and s is the variance of the study region.
9. The detection method according to claim 8, wherein s is calculated by the formula:
Figure FDA0002425652280000033
wherein, Δ vjRepresenting the difference between the historical average speed and the current average speed of path element j,
Figure FDA0002425652280000034
represents the average of all average velocity differences within the investigation region and n represents the total number of path elements within the investigation region.
10. The detection method according to claim 1, wherein the step 5 comprises:
step 5.1, respectively aiming at each candidate path, executing the following steps:
generating N simulation data sets for the candidate paths;
computing a likelihood ratio statistic score LL R for the candidate paths in each simulated datasetobs；
Through a male
Figure FDA0002425652280000035
Calculating the significance of the candidate path; wherein, # (f)i) Indicates that the condition f is satisfiediN is the total number of simulation data sets, LL RresScoring a likelihood ratio statistic for said candidate paths in the real data, SiRepresents said candidate path, p _ value (S)i) Representing the significance of the candidate path;
when the significance p _ value of the candidate path (S)i) And when the significance of the candidate route is less than or equal to α, determining that the significance of the candidate route meets a preset significance level, and determining that the candidate route is a sporadic congestion route, wherein α is the preset significance level.
CN202010219735.9A 2020-03-25 2020-03-25 Detection method for sporadic congestion path Active CN111402585B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010219735.9A CN111402585B (en) 2020-03-25 2020-03-25 Detection method for sporadic congestion path

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010219735.9A CN111402585B (en) 2020-03-25 2020-03-25 Detection method for sporadic congestion path

Publications (2)

Publication Number Publication Date
CN111402585A true CN111402585A (en) 2020-07-10
CN111402585B CN111402585B (en) 2021-02-02

Family

ID=71414031

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010219735.9A Active CN111402585B (en) 2020-03-25 2020-03-25 Detection method for sporadic congestion path

Country Status (1)

Country Link
CN (1) CN111402585B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113793692A (en) * 2021-09-17 2021-12-14 中南大学 Detection method for respiratory infectious disease infection source region
CN114550451A (en) * 2022-02-18 2022-05-27 平安国际智慧城市科技股份有限公司 Vehicle congestion early warning method, device, equipment and storage medium for parking lot

Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101727747A (en) * 2009-12-16 2010-06-09 南京信息工程大学 Abnormal road jam alarming method based on flow detection
CN101751777A (en) * 2008-12-02 2010-06-23 同济大学 Dynamic urban road network traffic zone partitioning method based on space cluster analysis
CN104809878A (en) * 2015-05-14 2015-07-29 重庆大学 Method for detecting abnormal condition of urban road traffic by utilizing GPS (Global Positioning System) data of public buses
CN104933240A (en) * 2015-06-10 2015-09-23 中国人民解放军装甲兵工程学院 Optimum design method for layout of armored vehicle cooling system
CN105389982A (en) * 2015-11-15 2016-03-09 安徽科力信息产业有限责任公司 Method for detecting and dispersing city intersection jam event based on FCD
CN106971537A (en) * 2017-04-20 2017-07-21 山东高速信息工程有限公司 For the congestion in road Forecasting Methodology and system of accident
CN107239506A (en) * 2017-05-11 2017-10-10 中国地质大学（武汉） A kind of autocorrelative appraisal procedure of geographic event space-time
CN107610469A (en) * 2017-10-13 2018-01-19 北京工业大学 A kind of day dimension regional traffic index forecasting method for considering multifactor impact
CN107958302A (en) * 2017-11-17 2018-04-24 中南大学 Empirical path planing method based on virtual topology transportation network
CN107978153A (en) * 2017-11-29 2018-05-01 北京航空航天大学 A kind of multimode traffic factors influencing demand analysis method based on space vector autoregression model
WO2018228279A1 (en) * 2017-06-12 2018-12-20 Beijing Didi Infinity Technology And Development Co., Ltd. Systems and methods for analyzing and adjusting road conditions
CN110826429A (en) * 2019-10-22 2020-02-21 北京邮电大学 Scenic spot video-based method and system for automatically monitoring travel emergency

Patent Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101751777A (en) * 2008-12-02 2010-06-23 同济大学 Dynamic urban road network traffic zone partitioning method based on space cluster analysis
CN101727747A (en) * 2009-12-16 2010-06-09 南京信息工程大学 Abnormal road jam alarming method based on flow detection
CN104809878A (en) * 2015-05-14 2015-07-29 重庆大学 Method for detecting abnormal condition of urban road traffic by utilizing GPS (Global Positioning System) data of public buses
CN104933240A (en) * 2015-06-10 2015-09-23 中国人民解放军装甲兵工程学院 Optimum design method for layout of armored vehicle cooling system
CN105389982A (en) * 2015-11-15 2016-03-09 安徽科力信息产业有限责任公司 Method for detecting and dispersing city intersection jam event based on FCD
CN106971537A (en) * 2017-04-20 2017-07-21 山东高速信息工程有限公司 For the congestion in road Forecasting Methodology and system of accident
CN107239506A (en) * 2017-05-11 2017-10-10 中国地质大学（武汉） A kind of autocorrelative appraisal procedure of geographic event space-time
WO2018228279A1 (en) * 2017-06-12 2018-12-20 Beijing Didi Infinity Technology And Development Co., Ltd. Systems and methods for analyzing and adjusting road conditions
CN107610469A (en) * 2017-10-13 2018-01-19 北京工业大学 A kind of day dimension regional traffic index forecasting method for considering multifactor impact
CN107958302A (en) * 2017-11-17 2018-04-24 中南大学 Empirical path planing method based on virtual topology transportation network
CN107978153A (en) * 2017-11-29 2018-05-01 北京航空航天大学 A kind of multimode traffic factors influencing demand analysis method based on space vector autoregression model
CN110826429A (en) * 2019-10-22 2020-02-21 北京邮电大学 Scenic spot video-based method and system for automatically monitoring travel emergency

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
ZIHAN KAN ETC.: "Fine-grained analysis on fuel-consumption and emission from", 《JOURNAL OF CLEANER PRODUCTION》 *
崔德冠: "基于公交GPS数据的城市道路偶发性拥堵检测与系统实现", 《中国优秀硕士学位论文全文数据库》 *
邓敏等: "基于泛在位置数据的城市道路网精细建模", 《中南大学学报》 *
高林等: "基于互联网信息的城市道路偶发拥堵判别算法研究与应用", 《工业仪表与自动化装置》 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113793692A (en) * 2021-09-17 2021-12-14 中南大学 Detection method for respiratory infectious disease infection source region
CN113793692B (en) * 2021-09-17 2023-10-17 中南大学 Method for detecting infectious source area of respiratory tract infectious disease
CN114550451A (en) * 2022-02-18 2022-05-27 平安国际智慧城市科技股份有限公司 Vehicle congestion early warning method, device, equipment and storage medium for parking lot
CN114550451B (en) * 2022-02-18 2023-08-18 平安国际智慧城市科技股份有限公司 Vehicle congestion early warning method, device and equipment for parking lot and storage medium

Also Published As

Publication number Publication date
CN111402585B (en) 2021-02-02

Similar Documents

Publication Publication Date Title
Autey et al. Safety evaluation of right-turn smart channels using automated traffic conflict analysis
CN109544932B (en) Urban road network flow estimation method based on fusion of taxi GPS data and gate data
CN104574967B (en) A kind of city based on Big Dipper large area road grid traffic cognitive method
WO2022166239A1 (en) Vehicle travel scheme planning method and apparatus, and storage medium
CN100357987C (en) Method for obtaining average speed of city rode traffic low region
CN104021670B (en) A method for extracting vehicle queuing status information in urban road network from high-resolution remote sensing images
CN108320511A (en) Urban highway traffic sub-area division method based on spectral clustering
CN105825669A (en) System and method for identifying urban expressway traffic bottlenecks
JP5424754B2 (en) Link travel time calculation device and program
CN104615897B (en) Estimation method of road segment travel time based on low-frequency GPS data
CN106251642B (en) A kind of public transport road chain speed calculation method based on real-time bus GPS data
CN106097717B (en) Signalized intersections based on the fusion of two class floating car datas are averaged transit time method of estimation
CN106227859B (en) The method of the vehicles is identified from GPS data
US20220084400A1 (en) Green wave speed determination method, electronic device and storage medium
CN111402585B (en) Detection method for sporadic congestion path
CN107424410A (en) A kind of accident detection method calculated based on route travel time
CN112862204A (en) Path planning method, system, computer equipment and readable storage medium
Kinoshita et al. Traffic Incident Detection Using Probabilistic Topic Model.
CN111311910A (en) Abnormal track detection method for multi-level road-level floating vehicle
CN109935076A (en) A method for identifying frequent traffic bottlenecks on urban expressways
CN113942526A (en) Acceptable risk based automatic driving overtaking track planning method
CN114676917B (en) A method and system for evaluating the spatial distribution of empty taxis
Azizi et al. Estimation of freeway platooning measures using surrogate measures based on connected vehicle data
CN113538902B (en) A traffic state-based restoration method for vehicle trajectory data at intersections
CN106289036B (en) A Road Width Measurement Method Based on Floating Car Data Analysis

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant